DroneLoad

Aim of the project

This project aims to create a drone with the following abilities:

• Going through a window
• Going through a tunnel
• Scanning QR codes
• Picking up objects and delivering them

During the competition, a lot of reward points are given for all the tasks that are performed auntonomously. Last year, none of the teams did any king of automation, all the drones were driven by hand.

Developped algorithms

Currently, we have developped several algorithms, that are still to be implemented on the on bord computer (Jetson Nano):

• Autonomous window detection, optimal path computation and crossing (this can easily be adapted to the tunnel crossing)
• QR code reading
• Circles detection (in order to recognise in a picture the objects that are to be picked and delivered)

Package installation

pip install -r requirements

Project usage

cd to the destination of the main.py file

QRCode Reading

python main.py qrcode

Press 'q' to exit

Cirle detection

python main.py circles

Press 'q' to exit

Contours detection

python main.py contours

Press 'q' to exit

Rectangle detection

• In image issued from your camera :

python main.py vrect

Press 'q' to exit

• In preloaded image :

python main.py imrect path_to_image

Press 'q' to exit

Find path to window

• From predefined parameters :

python main.py path

Press 'q' to exit

• From preloaded image

python main.py impath path_to_image

Press 'q' to exit

• Animate calculated path (ATTENTION PROBLEME A REGLER)

python main.py vpath

Press 'q' to exit